Yahoo Big Thinkers Series – Automated Economy and Prediction market

13 Jun

At the 2nd of the lectures in the Yahoo Big Thinker Series, which happened yesterday at the Hotel Leela, Dr. David Pennock talked on how the internet and search engines have developed and are developing to make the economy more automated.

David is Principal Research Scientist at Yahoo. More about David is here:

What David discussed was about how search engines have evolved from being just word searchers to being auctioneers for what people are searching.

There are people who are willing to sell products, services and information, other people are searching. One of the key areas that David talks about is Continuous Double Auction.

Prediction market utilising CDA was also discussed. gives insurance to customer against bad weather, and the cost of which is calculated based on the prediction that the customer does. There are levels of severity of prediction that can be chosen from, and the website provides options to calibrate your prediction. is another success e-commerce that is based on predictions. You trade the predictions about events happening around the world, you can buy and sell them based on what your current take on them is. To know about how it works ->

A variety of questions were raised as to would this really work, if someone trades on a prediction like
Would terrorists attack on ___ time?“.
Such kind of predcitions themselves are not worth enough to trade about.

But the amount of intel such a model generates is enormous. Karthikeyan Iyer ( Director, Crafitti Consulting Pvt Ltd, ) asked me how would one monetize such kind of a model.

The prediction market is generating info about what is the current mind set of people and whether the prediction that is in hand is on the right lines of not.

Take an example : A hybrid car which generates more power than it
consumes will be ready in an economically viable state by 2015

The amount of response to this prediction gives a rough estimate how many people like this idea. If the P-value is > 50% it means there are fair chances of people buying such a product and companies like Toyota, Honda can basically utilise this intel to know where there target markets are.

Take another example : A Red coloured shampoo which is not harmful,
if eaten accidentally, will be a big hit

Now clearly an example of radically different idea that can be worked upon, especially from the point of view of children. But if the market is not positive about this prediction, this might be because of any of the things -> color, edible nature etc. This is a simple example, but complex predictions can be formulated to make sure that the intel generated is good enough to find out if a particular product is viable or not.


Point to ponder :

How prediction market concept can be utlized inside organisations to improve productivity of processes as well as employees?


See the image at the top of the post.
Me and my friend Vishal were discussing about the various interpretations of it when this is what came up:

How big can YOUTH INK?


3 Responses to “Yahoo Big Thinkers Series – Automated Economy and Prediction market”

  1. Abhinav Goyal June 13, 2009 at 11:02 pm #

    Hi Nitin,

    Firstly thanks for this useful post. Prediction concept has been used since our ancient India. People used to predict weather, time seeing the shadows of the sun etc. In modern India a very good example is derivatives. As such, prediction has been a very succesful business model for generating money as your examples also proove weatherbill and intrade.
    Coming how one can monetize such model
    1. See A can predict but B will agree only if A’s prediction has the probability of being false.
    2. How much less probability B takes now, depends upon his risk appetite.
    3. So firstly one has to measure his risk appetite before starting this model.
    4. This model cannot work only with few people bidding. This model shall be successful if there are large no of people predicting and betting.
    5. Can I say this model is similar to insurance model?

    I have the following doubts.

    1. what does CDA mean in your post?
    2. The topic was automated economy, how has he linked this to prediction concept business, its just a type of business…………… what special about automation here?
    3. Point to ponder: can we auction on the results that google will fetch?

    Do reply

    Best Regards,
    Abhinav Goyal

    • Nitin Gaopande June 14, 2009 at 9:15 am #

      Dear Abhinav,
      1. what does CDA mean in your post?
      – CDA is continuous double auction as i mentioned in the post. You can find more here :
      CDA is an old concept where buying and selling part of the action happens independently till the prices on both the sides are tying up…

      2. The topic was automated economy, how has he linked this to prediction concept business, its just a type of business…………… what special about automation here?
      – By automated economy, one means how the transactions that happen in the economic world are getting more and more automated. A manual teller is replaced by an automated teller (ATM) in the banking world. Same way, auctions are now become lot more online with computers taking place of the actioneer.

      3. Point to ponder: can we auction on the results that google will fetch?
      – Thats the business model of Google, in fact, more or less, it itself auctions automatedly the position of the search and the hits it gets based on the position. Google’s whole rev model is based on the fact that people want to hit something that appeals to them in their first look on the first page in a quick go through.

  2. Abhinav Goyal June 14, 2009 at 10:32 pm #

    thanx for the information

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: